diff --git a/SYCL/Plugin/level_zero_queue_profiling.cpp b/SYCL/Plugin/level_zero_queue_profiling.cpp
new file mode 100644
index 0000000000..70600fc7b2
--- /dev/null
+++ b/SYCL/Plugin/level_zero_queue_profiling.cpp
@@ -0,0 +1,69 @@
+// REQUIRES: gpu, level_zero
+
+// RUN: %clangxx -fsycl -fsycl-targets=%sycl_triple  %s -o %t.out
+// RUN: env ZE_DEBUG=-1 %GPU_RUN_PLACEHOLDER %t.out 2>&1 | FileCheck --check-prefixes=WITHOUT %s
+// RUN: env ZE_DEBUG=-1 %GPU_RUN_PLACEHOLDER %t.out profile 2>&1 | FileCheck --check-prefixes=WITH %s
+
+// Test case adapted from the SYCL version of Rodinia benchmark hotspot.
+
+// clang-format off
+// Check the expected output when queue::enable_profiling is not specified
+//
+// WITHOUT: ze_event_pool_desc_t flags set to: 1
+// WITHOUT: SYCL exception caught: Native API failed. Native API returns: -7 (CL_PROFILING_INFO_NOT_AVAILABLE)
+
+// Check the expected output when queue::enable_profiling is specified
+//
+// WITH: ze_event_pool_desc_t flags set to: 5
+// WITH: Device kernel time:
+// clang-format on
+//
+
+#include <CL/sycl.hpp>
+using namespace cl::sycl;
+
+int foo(queue &q, int n) {
+  for (int i = 0; i < n; i++) {
+
+    sycl::event queue_event = q.submit([&](handler &cgh) {
+      cgh.parallel_for<class empty>(range<2>(10000, 10000),
+                                    [=](item<2> item) {});
+    });
+
+    q.wait();
+
+    // Get kernel computation time
+    try {
+      auto startk = queue_event.template get_profiling_info<
+          cl::sycl::info::event_profiling::command_start>();
+      auto endk = queue_event.template get_profiling_info<
+          cl::sycl::info::event_profiling::command_end>();
+      auto kernel_time =
+          (float)(endk - startk) * 1e-9f; // to seconds, 1e-6f to milliseconds
+      printf("Device kernel time: %.12fs\n", (float)kernel_time);
+
+    } catch (const sycl::exception &e) {
+      std::cout << "SYCL exception caught: " << e.what() << '\n';
+      return 0;
+    }
+  }
+  return n;
+}
+
+int main(int argc, char **argv) {
+
+  bool profiling = argc > 1;
+
+  {
+    gpu_selector dev_sel;
+    property_list propList{};
+    if (profiling)
+      propList = cl::sycl::property::queue::enable_profiling();
+
+    queue q(dev_sel, propList);
+    // Perform the computation
+    foo(q, 2);
+  } // SYCL scope
+
+  return 0;
+}